Stack expects a non-empty TensorList #101

New Issue

mocandragon · 2023-03-09T07:20:32Z

mocandragon commented

2023-03-09 07:20:32 +00:00

Hi!
I installed this a few days ago and tried to generate something, but got a CUDA out of memory error. Now, a few days later, I've updated and runt he force update batch file, and now I'm getting a different error

File "...\ai-voice-cloning\modules\tortoise-tts\tortoise\api.py", line 431, in get_conditioning_latents
    auto_conds = torch.stack(auto_conds, dim=1)
RuntimeError: stack expects a non-empty TensorList

Is there anything I can do to fix this?
Thanks!

Hi! I installed this a few days ago and tried to generate something, but got a CUDA out of memory error. Now, a few days later, I've updated and runt he force update batch file, and now I'm getting a different error ``` File "...\ai-voice-cloning\modules\tortoise-tts\tortoise\api.py", line 431, in get_conditioning_latents auto_conds = torch.stack(auto_conds, dim=1) RuntimeError: stack expects a non-empty TensorList ``` Is there anything I can do to fix this? Thanks!

mrq commented

2023-03-09 13:07:21 +00:00

Insufficient stack trace.

mrq added the

insufficient info

label 2023-03-09 19:13:11 +00:00

mrq commented

2023-03-11 16:47:19 +00:00

Ah, I managed to have it happen myself. Should be remedied in commit cf41492f76. If not:

at least for me, it cropped up from having a dataset prepared, but didn't load anything (for example, if the accompanied train.txt is empty)
set a voice chunk size of 1

Ah, I managed to have it happen myself. Should be remedied in commit cf41492f76d2bb93cb2972bb5585edcf10d531ca. If not: * at least for me, it cropped up from having a dataset prepared, but didn't load anything (for example, if the accompanied `train.txt` is empty) * set a voice chunk size of 1

mrq closed this issue

2023-03-11 16:47:19 +00:00

Sign in to join this conversation.